Most of us wouldn’t think there was anything problematic about the meaning of names. When I use the name ‘Brad’ to talk about my good friend, we wouldn’t think to question what I actually meant by the name ‘Brad’. After all, I just mean Brad – that guy who pops up on MSN chat to distract me from my studies. But it turns out that a workable semantics of names is one of the most elusive goals of the philosophy of language today. In a very real sense, we still don’t know what names mean. There is still no agreement on how it is that we manage to use names as we do, or even how it is that we can refer to objects at all. At this point, your commonsense ideas are probably causing you to scoff. Of course we know what names mean – they mean whatever object they stand for! And maybe they do. By the end of this article, you might still think as you did at the beginning – but I doubt you’ll remain so confident.

The Direct Reference View

The commonsense view – once articulated by John Stuart Mill – is that the meaning of a name is its referent. But what does this mean? This doesn’t tell us very much at all unless we situate this claim within a broader semantic framework. After all, names are most commonly found in sentences, and sentences as a whole can be said to have meaning too. What we want to say then, is that names contribute their referent to the meaning of a sentence. But how do we articulate what the meaning of sentences are? And how does the contribution of a name determine, at least in part, the meaning of a whole sentence?

Perhaps the most popular view of the twentieth century, defended by philosophical giants as Frege and many after him, is known as the truth conditional view of meaning. The basic idea is that you understand a sentence when you understand the conditions of its truth, i.e. you understand what in the world has to take place that would cause us to view that sentence as being true. In the context of this theory of meaning, the direct reference view states that a name contributes its reference to the truth conditions of the sentence. To get a little clearer on what this means, consider the following example sentence:

1) Brad is logged on to MSN chat.

This sentence is true just when Brad is indeed logged on to MSN. We say that the name “Brad” contributes the object [Brad] because it doesn’t matter to the truth of this sentence whether anyone else is logged in – only Brad. The predicate section of the sentence (the bit that says: “is logged on to MSN chat”) makes a contribution as well. Rather than contributing an object, it contributes a function. Think of a mathematical function like addition. To get some kind of output from that function, you would plug in two numbers, and get another number as an output. Predicates work in much the same fashion, but instead of plugging in numbers, you plug in an object. And instead of getting a number as an output, you get a truth value (as determined by the truth conditions and the state of the world).

So we have a clearer idea now of how it is that names are supposed to contribute to the meaning of a sentence as a whole. But alas things are not so simple. There are three classic problems which historically have caused great problems for this view.

The Problem of Bearerless Names

Given the title of this problem you’ve probably already guessed what it’s going to be. Consider the sentence:

2) Santa Claus is fat.

Now one problem we might have is that we have no clear intuitions as to whether this sentence is true or not. Some might say it isn’t, but some might say, that in a sense it is true. This is not because we have differing information about the world, its not that some of us think that Santa Claus is thin, but because we aren’t entirely sure what the truth conditions are when the object in question doesn’t actually exist. With sentence 1) – there was no problem, the object was there to plug into the function to yield our truth value. But in this case there is no object that exists to actually plug into the predicate function. But then, under the direct reference view, combined with a truth conditional semantics, this sentence should be meaningless – yet it is clearly not meaningless.

Besides this, we have all sorts of intuitions as to the meaning of the name ‘Santa Claus’ – he’s that jolly, fat fellow that brings us presents on Christmas day – a fictional character, myth and legend – etc. But none of these intuitions correspond with an actual object that the direct reference view requires.

A more acute version of this problem can be seen in the following example:

3) Santa Claus does not exist.

Now we want this sentence to come out as true. But in order for it to come out as true, we need an object to plug into the function denoted by the predicate “does not exist”. But if that object exists to plug into the function, then it must exist! We are left with a paradoxical situation. This problem is known as the problem of negative existentials.

Frege’s Puzzle

This problem is sometimes referred to as ‘the problem of informativeness’ – and was discovered by Gottleb Frege, the famous logician. Consider the two names, ‘Hesperus’ and ‘Phosphorus’. These were the names the ancient Greeks used to refer to the evening star and the morning star respectively. Now we both know that both the evening star and the morning star are in fact the same object – the planet venus. But this wasn’t known back then – so when someone claimed that:

4) Hesperus is Phosphorus

this was a new claim – its truth represented new knowledge that we didn’t have before.

But the direct reference view of names makes it difficult to explain how this is possible. Given that both ‘Hesperus’ and ‘Phosphorus’ refer to the same object we should be able to make a substitution without change of meaning – as in the following example:

5) Hesperus is Hesperus

Under the direct reference view, sentence 5 should be equivalent in meaning to sentence 4 – but 4 is informative, while 5 is clearly not. Hence, it is argued, the direct reference view must have something wrong with it.

The Problem of Belief

Consider the following sentence:

6) Lois Lane believes that Superman is strong

Just pretend that this isn’t a fictional example – that all names in the sentence have real world referents. Given what we know about Superman and Lois Lane, we would judge this sentence to be true. But now consider the following substitution instance:

7) Lois Lane believes that Clark Kent is strong

Like in the previous puzzle, we have a failed substition instance. Lois Lane doesn’t believe that Clark Kent is strong – she only believes that Superman is strong. Under the direct reference view, however, this substitution instance should be valid. Examples like this seem to suggest that there is more to the meaning of a name than just the referent.

As it turns out, for direct reference theorists, the problem of belief contexts is perhaps the most intractable of the the three problems and still causes headaches today.

Because of these problems, philosophers were led to consider other theories throughout the twentieth century. The leading alternative was known as descriptivism, which was dominant up until Saul Kripke gave his famous ‘Naming and Necessity’ lectures in the 70s and reinstated direct reference as the dominant view. In my next post I’ll give an account of the descriptivist theory of names, and show how Kripke managed to put it to bed. In the meantime, I’ll leave you to try and figure some responses to the above three problems.

(Extra credit to those who know why I used a rose for the picture).